OSDN Git Service

perf test BPF: Fixup BPF test using epoll_pwait syscall function probe
authorArnaldo Carvalho de Melo <acme@redhat.com>
Wed, 18 Apr 2018 15:59:03 +0000 (12:59 -0300)
committerArnaldo Carvalho de Melo <acme@redhat.com>
Wed, 18 Apr 2018 18:35:53 +0000 (15:35 -0300)
commiteccb1b936363c62544bccb5bbb75afec9536f7e3
treead9b8aab29d322d65e5fab662cc11755132c318b
parent518c6021e9b82696819b380cd173e76cac55a01e
perf test BPF: Fixup BPF test using epoll_pwait syscall function probe

Since e145242ea0df ("syscalls/core, syscalls/x86: Clean up syscall stub
naming convention") changed the main syscall function for 'epoll_pwait'
to something other than the expected 'SyS_epoll_pwait the' 'perf test
BPF' entries started failing, fix it by using something called from the
main syscall function instead, 'epoll_wait', which should keep this test
working in older kernels too.

Before:

  # perf test BPF
  40: BPF filter                           :
  40.1: Basic BPF filtering                : FAILED!
  40.2: BPF pinning                        : Skip
  40.3: BPF prologue generation            : Skip
  40.4: BPF relocation checker             : Skip

If we use -v for that test we see the problem:

    Probe point 'SyS_epoll_pwait' not found.

After:

  # perf test BPF
  40: BPF filter                           :
  40.1: Basic BPF filtering                : Ok
  40.2: BPF pinning                        : Ok
  40.3: BPF prologue generation            : Ok
  40.4: BPF relocation checker             : Ok
  #

Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Dominik Brodowski <linux@dominikbrodowski.net>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Wang Nan <wangnan0@huawei.com>
Link: https://lkml.kernel.org/r/tip-y24nmn70cs2am8jh4i344dng@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
tools/perf/tests/bpf-script-example.c